Interface refinement in reactive systems

7Citations
Citations of this article
3Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Suppose one has a system that has a synchronous interface with its environment. Now, suppose that one refines this system and changes its interface to an asynchronous one. Whatever is meant here by refinement, it cannot be standard (process) refinement since the interface actions have changed; nor is it action refinement in the sense that a process is substituted for an action, as the intention presumably is to allow the system to proceed without having to wait until the environment is willing to synchronize. In this paper we propose a notion of interface refinement of which changing synchronous to asynchronous communication is an instance; as is in fact the reverse change. This notion of interface refinement is quite powerful; it generalizes all existing methods w.r.t, the class of interface changes that it allows. The major part of the paper is concerned with developing proof rules with which to verify interface refinement. We use (finear) temporal logic as specification language and an adaptation of the Manna-Pnueli verification framework. The method is illustrated by verifying an interface change in which synchronous communication is replaced by asynchronous send and receive. Proofs of the various theorems and lemma’s are delegated to an appendix.

Cite

CITATION STYLE

APA

Gerth, R., Kuiper, R., & Segers, J. (1992). Interface refinement in reactive systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 630 LNCS, pp. 77–93). Springer Verlag. https://doi.org/10.1007/bfb0084784

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free